WO2020073132A1

WO2020073132A1 - Skewed x-ray detection apparatus and method for pipeline use

Info

Publication number: WO2020073132A1
Application number: PCT/CA2019/051449
Authority: WO
Inventors: Stephen Geoffrey DRAKE; Simon FOWLER-DIXON; Stephen Knight
Original assignee: Shawcor Ltd.
Priority date: 2018-10-11
Filing date: 2019-10-10
Publication date: 2020-04-16

Abstract

Apparatus and methods for X-ray detection are disclosed that are suitable for use in scanning pipeline welds. The techniques involve the use of an array of X-ray sensitive tiles that are skewed relative to the scanning direction. The skew of the array eliminates blind spots at the gaps between tiles and has several other potentially advantageous features, particularly when used with highly sensitive tiles, such as Cadmium Telluride (CdTe) or Cadmium Zinc Telluride (CdZnTe) crystal arrays. Additional features are described for protecting the highly sensitive tiles from environmental conditions, such as extreme temperatures and kinetic shocks.

Description

SKEWED X-RAY DETECTION APPARATUS AND METHOD FOR PIPELINE USE

FIELD

[0001] The present disclosure is related to the field of X-ray girth weld inspection, in particular for oil and gas pipelines.

BACKGROUND

[0002] An oil and gas pipeline is typically made from a plurality of lengths of coated pipe joined together. The lengths of pipe are typically coated for insulation, impact resistance, water proofing, and corrosion resistance. The coating is often a quite thick layer on top of the steel pipe length. To facilitate welding the pipe lengths together, the coating ends before the steel end of the pipe, leaving a region, often about 24" wide, of exposed steel pipe, at the ends of the pipe before the coating begins. This is known as the "cutback region".

[0003] Pipe lengths are typically welded together in the field, utilizing what is called a "girth weld" - a weld around the perimeter of the steel pipe. Then the cutback region is covered or filled (or both) to provide insulation, impact resistance, water proofing, and corrosion resistance equal to the rest of the pipe coating. As can be appreciated, covering or filling the cutback region in the field is a much more expensive process than coating the pipe at the factory, and typically involves more expensive materials, but a much smaller length of pipe.

[0004] Best practices dictate that each girth weld must be inspected to ensure it meets certain quality criteria, before the cutback is filled. There are a variety of methods for girth weld inspection, but one of the most common is an X-ray inspection. A metal band is attached around the pipe and an X-ray detector assembly is attached to the band. The scanner assembly, which contains at least one X-ray detector, rotates around the band and scans the weld. The X-ray detector may inspect either or both of the weld sections proximal to the outer diameter and the inner diameter of the pipe. The cutback region must be wide enough to accommodate the detector and band during the inspection.

[0005] Conventionally, pipelines are inspected by penetrating the weld with X- rays or Gamma rays and collecting the resultant 'shadowgraph' image on an X- ray/gamma ray sensitive film. The film requires time consuming and

environmentally un-friendly chemical processing, washing and drying prior to viewing and storage.

[0006] In the case of pipelines where internal access is difficult to obtain, a known technique is to use a high-strength, broad-beam radioactive source to penetrate both walls of the pipeline to expose an X-ray/gamma ray sensitive film plate on the adjacent side of the pipeline. In many cases, up to six exposures are required to inspect the weld at all positions around the pipeline circumference. Even when using high strength radioisotope sources, the exposure times are long and the radiation exclusion area is extensive. Also, road transport and terrorist threats make the use of high strength radioisotope sources undesirable.

[0007] Digital X-ray technology exists to replace film radiography and Gamma sources, but these systems lack sensitivity and so require long scan times or a cumbersome, high-power X-ray source. The lack of sensitivity to X-rays of these existing detectors limits this technology for use with an X-ray source internal to the pipe or a fully externally X-ray source/detector on smaller-diameter, thinner-walled pipes.

[0008] When the existing digital X-ray systems are used to perform a fully external scan of the entire circumference of the pipe girth weld, they use a radiographic technique commonly known as 'Double Wall, Single Image' (DWSI). Such systems can take over one hour to scan a 36" heavy wall pipeline girth weld.

[0009] There thus exists an unmet need for a fast X-ray scanning technique for pipe girth welds that yields high-quality images.

SUMMARY OF THE INVENTION [0010] The present disclosure describes an apparatus and method for X-ray detection suitable for use in scanning pipeline welds. The technique involves the use of an array of X-ray sensitive tiles that are skewed relative to the scanning direction. The skew of the array eliminates blind spots at the gaps between tiles and has several other potentially advantageous features. When used with highly sensitive tiles, such as Cadmium Telluride (CdTe) or Cadmium Zinc Telluride

(CdZnTe) crystal arrays, this technique may allow for faster, higher-resolution scanning than conventional techniques. Additional features are described for protecting the highly sensitive tiles from environmental conditions, such as extreme temperatures and kinetic shocks.

[0011] According to some aspects, the present disclosure describes an X-ray detection apparatus for scanning a scanned object along a scan direction using X- ray detection. The apparatus comprises an X-ray detector and electronics module. The X-ray detector and electronics module comprises an X-ray sensitive area for detecting X-rays passing through or reflected from the scanned object, The X-ray sensitive area comprises a plurality of X-ray sensitive tiles. Each X-ray sensitive tile comprising a two-dimensional array of pixels. The plurality of X-ray sensitive tiles is arranged in a linear array having a width dimension and a height dimension perpendicular to the width dimension, the height dimension being skewed relative to the scan direction by a skew angle. The skew angle is greater than zero degrees and less than forty-five degrees. The X-ray detector and electronics module also comprises electronics for collecting X-ray image data from the X-ray sensitive area.

[0012] According to a further aspect which can be combined with other embodiments disclosed herein, the apparatus further comprises an actuator for driving the apparatus in the scan direction.

[0013] According to a further aspect which can be combined with other embodiments disclosed herein, each of the plurality of X-ray sensitive tiles comprises a crystal array comprising either Cadmium Telluride (CdTe) or Cadmium Zinc Telluride (CdZnTe). [0014] According to a further aspect which can be combined with other embodiments disclosed herein, each of the plurality of X-ray sensitive tiles further comprises a complementary metal-oxide-semiconductor 2-dimensional array bump- bonded to the crystal array.

[0015] According to a further aspect which can be combined with other embodiments disclosed herein, the apparatus further comprises an environmental housing for protecting the X-ray detector and electronics module from

environmental hazards.

[0016] According to a further aspect which can be combined with other embodiments disclosed herein, the apparatus further comprises a temperature control system for controlling the temperature inside the environmental housing, and wherein the environmental housing is insulated.

[0017] According to a further aspect which can be combined with other embodiments disclosed herein, the temperature control system comprises a Peltier device.

[0018] According to a further aspect which can be combined with other embodiments disclosed herein, the apparatus further comprises a shock absorbing suspension coupled between the X-ray detector and electronics module and the environmental housing for protecting the X-ray detector and electronics module against physical shocks.

[0019] According to a further aspect which can be combined with other embodiments disclosed herein, each X-ray sensitive tile comprises a rectangular matrix of pixels having a width dimension aligned with the width dimension of the linear array and a height dimension aligned with the height dimension of the linear array.

[0020] According to a further aspect which can be combined with other embodiments disclosed herein, the plurality of X-ray sensitive tiles are arranged in the linear array with a gap between each pair of adjacent tiles. [0021] According to a further aspect which can be combined with other embodiments disclosed herein, the apparatus further comprises a data processing module for receiving X-ray image data from the electronics.

[0022] According to a further aspect which can be combined with other embodiments disclosed herein, the apparatus further comprises a controller for perform a single scan by controlling the X-ray detector and electronics module to collect X-ray image data and provide this data to the data processing module, and controlling the actuator to move the apparatus a predetermined between-scan distance in the scan direction after each scan.

[0023] According to a further aspect which can be combined with other embodiments disclosed herein, the data processing module is configured to receive a first scan dataset from the X-ray detector and electronics module corresponding to the X-ray image data collected from a first scan, receive one or more additional scan datasets from the X-ray detector and electronics module corresponding to the X-ray image data collected from one or more scans following the first scan, de-skew the first scan dataset and the one or more additional scan datasets, and apply time delay integration to the first scan dataset and the one or more additional scan datasets to generate an output image.

According to a further aspect which can be combined with other embodiments disclosed herein, a portion of the output image corresponding to a portion of the scanned object is generated by applying time delay integration to a predefined number of datasets containing X-ray image data corresponding to the portion of the scanned object. Applying time delay integration to generate an output image comprises, once the predefined number of datasets corresponding to a first portion of the scanned object have been received by the data processing module, generating a first portion of the output image corresponding to the first portion of the scanned object by applying time delay integration to the predefined number of datasets; sending the first portion of the output image to a user device; and repeating the steps of generating a portion of the output image and sending the portion of the output image to the user device for each subsequent portion of the output image as subsequent datasets are received by the data processing module.

[0024] According to a further aspect which can be combined with other embodiments disclosed herein, the portion of the output image is a one-pixel-wide line of the output image.

[0025] According to a further aspect which can be combined with other embodiments disclosed herein, the between-scan distance is less than a distance corresponding to a single pixel height in the X-ray image data, and the data processing module is configured to receive a first scan dataset from the X-ray detector and electronics module corresponding to the X-ray image data collected from a first scan; receive one or more additional scan datasets from the X-ray detector and electronics module corresponding to the X-ray image data collected from one or more scans following the first scan; de-skew the first scan dataset and the one or more additional scan datasets; and apply time delay integration to the first scan dataset and the one or more additional scan datasets to generate an output image, wherein de-skewing a dataset comprises interpolating between adjacent pixels in the dataset to increase the number of pixels in the dataset.

[0026] According to a further aspect which can be combined with other embodiments disclosed herein, the electronics are configured to segment the X-ray image data into a plurality of energy bands.

[0027] According to a further aspect which can be combined with other embodiments disclosed herein, the electronics are configured to filter image noise in the X-ray image data based on an energy threshold.

[0028] According to a further aspect which can be combined with other embodiments disclosed herein, the present disclosure describes a method for scanning a scanned object along a scan direction using X-ray detection, comprising applying an X-ray source to the scanned object; detecting X-rays passing through or reflected from the scanned object using a plurality of X-ray sensitive tiles, each X-ray sensitive tile comprising a two-dimensional array of pixels, the plurality of X- ray sensitive tiles being arranged in a linear array having a width dimension and a height dimension perpendicular to the width dimension, the height dimension being skewed relative to the scan direction by a skew angle, the skew angle being greater than zero degrees and less than forty-five degrees; collecting X-ray image data from the X-ray sensitive tiles; moving the X-ray sensitive tiles a predetermined distance along the scan direction; and repeating the steps of applying, detecting, collecting, and moving until a region of interest of the scanned object has been scanned.

[0029] According to a further aspect which can be combined with other embodiments disclosed herein, each of the plurality of X-ray sensitive tiles comprises a complementary metal-oxide-semiconductor 2-dimensional array bump- bonded to a crystal array comprising either Cadmium Telluride (CdTe) or Cadmium Zinc Telluride (CdZnTe).

BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Reference will now be made, by way of example, to the accompanying drawings which show example embodiments of the present application, and in which :

[0031] FIG. 1 is a simplified top view of an X-ray detection apparatus according to a further example embodiment showing the angle of skew of the X-ray sensitive area of the apparatus relative to the scan direction of the apparatus.

[0032] FIG. 2A is a simplified top view of a pair of skewed X-ray detector tiles according to an example embodiment showing the motion of the weld being scanned relative to the area scanned by the tiles.

[0033] FIG. 2B is a simplified top view of a pair of conventional, non-skewed X-ray detector tiles showing the motion of the weld being scanned relative to the area scanned by the tiles. [0034] FIG. 3 is a simplified front view of an internal climate control system for an X-ray scanner apparatus according to an example embodiment.

[0035] FIG. 4A is a top view of an X-ray detector apparatus shock mounting assembly according to an example embodiment.

[0036] FIG. 4B is a front view of the X-ray detector apparatus shock mounting assembly of FIG. 4A.

[0037] FIG. 5 is a simplified top view of a six-tile X-ray detector array showing the angle of skew of the array relative to the scan direction of the array.

[0038] FIG. 6 is a flowchart showing an example method for time delay integration of skewed X-ray scanning data according to an example embodiment.

[0039] FIG. 7 is a flowchart showing an example method for increasing the image resolution of skewed X-ray scanning data by combining adjacent pixel data according to an example embodiment.

[0040] FIG. 8A is an example image generated by a non-skewed X-ray scanning apparatus.

[0041] FIG. 8B is an example image generated by a skewed X-ray scanning apparatus according to an example embodiment.

[0042] FIG. 9 is a perspective view of an example x-ray detector and x-ray source scanner attached to an example band allowing traversal of a pipe.

[0043] FIG. 10 is a perspective view of an example X-ray detector apparatus according to an example embodiment mounted on a track to scan a pipe weld.

[0044] FIG. 11 is a front view of an example X-ray detector apparatus according to an example embodiment mounted on a track to scan a pipe weld.

[0045] FIG. 12 is a close up front view of an example X-ray detector apparatus according to an example embodiment mounted on a track to scan a pipe weld. [0046] FIG. 13 is a front view of a pipe weld showing the scan path of an example X-ray detector apparatus and the implementation of time delay integration using a skewed configuration of the apparatus' X-ray detection array.

[0047] FIG. 14A is an example image generated by a non-skewed X-ray scanning apparatus, corresponding to the drawing in FIG. 8A.

[0048] FIG. 14B is an example image generated by a skewed X-ray scanning apparatus according to an example embodiment, corresponding to the drawing in FIG. 8B.

[0049] FIG. 15 is a photograph showing a perspective view of an example X- ray detector apparatus according to an example embodiment mounted on a track to scan a pipe weld, corresponding to the drawing in FIG. 10.

[0050] FIG. 16 is a photograph showing a front view of an example X-ray detector apparatus according to an example embodiment mounted on a track to scan a pipe weld, corresponding to the drawing in FIG. 11.

[0051] FIG. 17 is a photograph showing a close up front view of an example X-ray detector apparatus according to an example embodiment mounted on a track to scan a pipe weld, corresponding to the drawing in FIG. 12.

[0052] Similar reference numerals may have been used in different figures to denote similar components.

DESCRIPTION OF EXAMPLE EMBODIMENTS

[0053] The present disclosure relates to an apparatus and method for performing X-ray inspection of the integrity of girth welds in pipelines without the need to use an internal x-ray source or broad-beam external gamma ray source. In some examples, the apparatus is a highly sensitive inspection apparatus for digital real-time X-ray inspection. The described examples may eliminate the need to use X-ray/gamma ray film plates and their associated chemistry. In some examples, the performance of the apparatus and method exceeds the previous generation of x-ray detectors used for this application in sensitivity, inspection speed and resolution.

[0054] Described examples take advantage of recent developments in x-ray photon counting technology and overcomes the limitations of this technology, allowing it to be used in pipeline girth weld inspection. Due to these technical advances, some example detectors can show up to eight-fold increase in x-ray sensitivity compared to the previous generation of detectors.

[0055] Examples described herein are based on core x-ray detection

technology recently being made commercially available as of 2018. However, this base technology alone is not suited for the exacting requirements of pipeline weld inspection.

[0056] With reference to the drawings, FIG. 1 shows a simplified view of an X- ray detector apparatus 150. The apparatus 150 comprises an outer environmental housing 152 containing an X-ray detector and electronics module 158. The X-ray detector and electronics module 158 includes an X-ray sensitive area 154. In some embodiments, the X-ray sensitive area 154 is configured as a rectangular array 157 of X-ray sensitive tiles 155.

[0057] In one embodiment, the X-ray sensitive area 154 consists of a rectangular array 157 of six X-ray sensitive tiles 155 measuring 76mm in length total. In another embodiment, the X-ray sensitive area 154 consists of a

rectangular array 157 of eight X-ray sensitive tiles 155 measuring 102mm in length. Other embodiments may include a greater or lesser number of tiles and have a greater or lesser length.

[0058] The X-ray sensitive area 154 of the apparatus 150 is skewed at a skew angle 160 relative to the scanning direction 162 (and to line 153 parallel to the scanning direction 162). This skew serves to eliminate imaging gaps at the boundaries between individual tiles of the X-ray sensitive area 154, as described in greater detail below. [0059] The amount of skew is application dependent and may vary in different example embodiments, with common skew angles being between 3 degrees and 15 degrees.

[0060] In the context of a pipe weld scanning application, the scan direction 162 is typically in the direction of the weld, i.e. circumferentially around the pipe.

[0061] In some embodiments, the X-ray sensitive area 154 is made up of a series of X-ray sensitive modules or tiles 155. Each tile 155 is 128 pixels in width by 256 pixels high. The tiles 155 in some embodiments are complementary metal- oxide-semiconductor (CMOS) 2-dimensional arrays bump-bonded to a Cadmium Telluride (CdTe), or alternatively Cadmium Zinc Telluride (CdZnTe), crystal array. Each pixel has a 100-micron pitch. Therefore six tiles 155 can cover a 76mm weld width, or 8 tiles 155 can cover a 102mm weld width. Other array heights are possible, including 128 and 64 pixels.

[0062] Relative to prior module designs utilizing polycrystalline cesium iodide as an X-ray sensitive material, the use of CMOS 2-dimensional arrays bump-bonded to a CdTe or CdZnTe crystal array generally provides about twice the degree of X- ray sensitivity and a significant improvement in resolution due to the absence of cross-talk to adjacent pixels. Whereas prior cesium iodide designs worked by converting X-rays into photons to be detected by a photon detector, the

CdTe/CdZnTe crystals produce direct electrical effects within the CMOS in response to exposure to X-ray energy. This may allow CdTe/CdZnTe designs to implement desirable features such as segmentation of an image into different energy bands and filtering image noise based on an energy threshold.

[0063] FIG. 5 provides a more detailed view of the skewed configuration of an example X-ray sensitive area 154 of the apparatus 150, embodied as an array 157 of six X-ray sensitive tiles 155. In this embodiment, each tile 155 is 256 pixels in a height dimension 506 by 128 pixels in a width direction 504. The boundaries between the tiles 155 result in small gaps 203, as described in further detail below. The skew angle 160 from the horizontal 506 is, of course, identical to the skew angle 160 from the line 153 parallel to the scan direction 162 from FIG. 1, as the horizontal 506 in FIG. 5 is defined as perpendicular to the scan direction 162.

[0064] The X-ray modules or tiles 155 described above alone may not be entirely suited for weld inspection due to a number of potential problems with deployment of these tiles 155 in a conventional X-ray scanner using conventional imaging techniques.

[0065] The first potential problem is that there is a gap 203 between each pair of tiles 155. X-rays penetrating the weld and passing through these gaps are not detected. Weld features in these areas will not be imaged and will be missed.

[0066] This potential problem is addressed by the skewed configuration of the X-ray sensitive area 154. FIG. 2A and 2B show how this skew serves to capture images of weld features that would otherwise pass through the gap and escape imaging. In FIG. 2A, a first tile 202 and second tile 204 are shown. These tiles 202,204 represent adjacent X-ray sensitive tiles in an array making up the X-ray sensitive area 154 of an example apparatus 150 as in FIG. 1. The skew angle 160 of this apparatus 150 means that as the apparatus moves in the scan direction 162 relative to the pipe weld, a weld feature 206 may move along a first weld feature path 208, ensuring that the weld feature 206 passes across one or both of the first tile 202 and second tile 204 instead of passing through gap 203 between the tiles 202,204.

[0067] In contrast, FIG. 2B shows a possible failure mode of a scanning apparatus without a skewed detector. Here, the tiles 202,204 travel in a scan direction 162 parallel with the sides of the tiles, potentially resulting in a weld feature 206 travelling along a second weld feature path 210 that passes directly through the gap 203 between the tiles 202,204 and thereby evades detection or imaging.

[0068] A second potential problem with using the X-ray sensitive tiles 155 alone is that the Cadmium Telluride (CdTe) or Cadmium Zinc Telluride (CdZnTe) at the tile boundaries may have a different response to X-rays relative to the adjacent pixels, which may make the X-ray response unpredictable. This may cause artefacts to appear in the final image even after software corrections. These artefacts can be confused with weld defects such as cracks or lack of root fusion.

[0069] The skewed configuration of the apparatus 150 accounts for this potential problem as well. The array 157 of tiles 155 (e.g. six or eight tiles) is read- out as a single frame of pixels (e.g. 768 x 256 pixels for a six-tile array, 1024 x 256 pixels for an eight-tile array) into a data processing module 170, such as a control computer. Skewing the X-ray sensitive area 154 so that the gaps 203 are not aligned with the motion of the detector relative to the weld (along scan direction 162) allows all parts of the weld to be imaged.

[0070] In some embodiments, a single frame is read out for every 100 microns of surface scan of the weld, and 100 microns corresponds to 1 pixel. Each frame may be corrected for shading and bad pixels in real-time using the data processing module 170. Each frame is then de-skewed to 0 degrees orientation and stacked with the 255 preceding frames, each displaced by 1 pixel. Corresponding lines from each image are then summed or otherwise combined to give a single line of data. In an example embodiment, the corresponding lines of each of the images are summed, and each line therefore is increased in amplitude by 256 times. This results in the boundary anomalies being completely removed, having been dispersed over the 256 lines of data. The technique is referred to as 'Time Delay Integration' (TDI). In some embodiments, the single line of data generated by combining the lines from each image is provided to a user device, such as a computer terminal with a display, to be displayed to a user for review. This may happen in real-time as the device scans a weld: for example, once 256 images have been created using 256 scans, the first line of the TDI output can be generated and displayed to a user, with another line displayed after each subsequent scan as the device moves around the circumference of the pipe.

[0071] The skewed configuration of the array 157 requires adjustment of the conventional TDI technique. FIG. 13 shows how TDI may be implemented using a skewed array 157. A section of pipe 1302 is shown, including a pipe weld 1304 that is the subject of the scan. An example apparatus is configured to traverse the pipe 1302 circumferentially in scan direction 162, resulting in the example scan path defined by left border 1316 and right border 1318. Each scan of the apparatus results in a single rectangular image corresponding to the shape of the array 157 and its position relative to the pipe weld 1304, shown here as a series of skewed rectangular images beginning with top image 1306 and ending with bottom image 1308. (The figure is simplified for the sake of visibility - in the example

embodiments described herein, the number of overlapping images would be a large number such as 256, with each image being displaced along the scan direction 162 by a small distance, such as 100 microns. )In some embodiments, when the top image 1306 is captured, the data representing its left side corner 1312 and right side corner 1314 falling on the left and right side, respectively, of the left border 1316 and right border 1318 of the scan path, are discarded. Only the image data falling within the scan path defined by the left border 1316 and right border 1318 is retained, because this is the only data for which a sufficient number of overlapping images will be captured in the traversal of the weld 1304.

[0072] When a sufficient number of overlapping images has been captured to apply TDI to a portion of the pipe - in this case, when region 1310 has been captured in the full set of overlapping images from the top image 1306 to the bottom image 1308 - TDI may be applied to those overlapping images with respect to that region 1310 and action may be taken on that output, e.g. it may be sent to a user terminal to be viewed by a user. In this case, region 1310 will be a line having a length equal to the width of the array 157, foreshortened by the degree of the skew (i.e., foreshortened by the length of the lower segment of the left corner 1312). In some embodiments, the width of the scan path may be approximately 70mm, with each corner 1312,1314 protruding approximately 3mm to each side. These ratios will potentially vary with the amount of skew and the relative width : height ratio of the array 157.

[0073] In some embodiments, a single traversal of the pipe circumference 1302 is sufficient to generate usable data. In some embodiments, the apparatus may be configured to complete slightly more than a single traversal, such as a few extra degrees, to create overlap between the initial scans and the final scans. This overlap may account for imaging problems with the initial scans and for

irregularities in the pipe diameter, ensuring that no portion of the pipe

circumference goes un-scanned.

[0074] The collection of scan image data by the X-ray detector and electronics module 158 and/or the data processing module 170, as well as the movement of the apparatus 150 in the scan direction 162, are controlled in some embodiments by a controller. The controller may be implemented by the X-ray detector and electronics module 158 and/or the data processing module 170, and/or by a separate control element. The controller may be implemented by some combination of electronic circuits and software.

[0075] The movement of the apparatus in some embodiments is effected by an actuator, such as a motor, which may be configured in different embodiments to move the apparatus 150 along a track, band, or path using suitable means such as treads, gears, or wheels.

[0076] FIG. 6 is a flowchart showing an example TDI method 600 for collecting scanned image data for an object (such as a pipe weld) using a skewed array of sensor tiles (such as X-ray sensitive tiles 155 used as the X-ray sensitive area 154 of scanning apparatus 150, with a longitudinal axis of the array along which the tiles 155 are arranged being skewed relative to a scan direction 162). At step 602, the skewed array (such as array 157) collects a first image dataset from a first position relative to the object. The first image dataset comprises a

multidimensional array or matrix of pixels. This first image dataset is passed to a data processing module 170, such as a computer connected via a wired or wireless data connection to the scanning apparatus 150, which de-skews the image dataset to align it with the scan direction at step 604. At step 606 the skewed array is moved in the scan direction 162 a between-scan distance corresponding to a predetermined number of pixels (such as 100 microns, corresponding to one pixel). At step 608, a second image dataset is collected by the skewed array, and this second image dataset is passed to the data processing module 170 at step 610. Steps 606, 608 and 610 may then repeat zero or more times (such as N times, corresponding to the number of scan locations required to scan the object from beginning to end) to collect third and subsequent image datasets. At step 612, the data processing module processes the image datasets by applying a combination function to the datasets that combines pixel data from a plurality of the datasets where the pixel data in each such dataset corresponds to the same spatial location on the object being scanned. In one example embodiment, wherein the distance moved between each scan is 100 microns and this distance corresponds to one pixel, and wherein the height dimension of the image is 256 pixels (corresponding to a 256-pixel-high tile 155), the combination function operates by combining lines of pixels (along an image axis perpendicular to the scan direction) offset by one pixel between time-adjacent scans. Thus, the 1000^th line of the first image dataset would be combined with the 999^th line of the second image dataset, the 998^th line of the third image dataset, and so on through the 745^th line of the 256^th image dataset.

[0077] This combination function in some embodiments is an image value summing function applied to each pixel. In other embodiments, it may be an image value averaging function or a function that discards outlier pixel values from the set of pixels corresponding to that spatial location. Some embodiments may apply the combination function to groups of pixels rather than individual pixels. A skilled person will appreciate that a number of image-enhancement algorithms may be applied to a collection of images containing overlapping image data with multiple pixel samples corresponding to the same spatial or visual-field location.

[0078] In some embodiments, the data processing module 170 may stack the image datasets at an offset (such as a one-pixel offset) following de-skewing them at step 604, thereby simplifying the combination function at step 612. Furthermore, some embodiments may generate each single-line output using the TDI

combination function as soon as enough datasets have been collected to generate that line output. This single-line output can then be sent to an operator's display (in communication with the data processing module 170) for interpretation.

[0079] It is important to note that each image set contains image data from multiple tiles in the array, and the combination function will combine pixels collected by different tiles in different datasets. The skew of the array means that a fixed location on the pipe weld or other object that is located close to a gap 203 between tiles will be scanned in some datasets by a first tile 202 and in other dataset by a second tile 204, while it may not be included at all in some intervening datasets corresponding to scanning locations where it falls within the gap 203.

[0080] In some embodiments, image datasets may be generated for multiple passes across or around an object, with each pass displaced orthogonally to the scan direction. These passes may generate overlapping image data, and this overlap may result in corresponding pixels that may be used as part of the combination function.

[0081] A third potential problem with the X-ray sensitive tiles is that they may need to be kept at a constant temperature, typically within +/- 0.50C, for optimum performance. This makes them potentially unsuitable for the harsh environmental temperature ranges that are typically encountered in the field.

[0082] To address this problem, example embodiments may include an outer environmental housing 152 as shown in FIG. 1 in simplified form and in more detail in FIG. 3. In the example temperature control system 300 shown in front view in

FIG. 3, the X-ray detector and electronics module 316, including the detection tiles

155 and their read-out electronics, are encased in an external environmental casing

318, made from a highly insulative material. The interior of the housing or casing

318 is kept at a constant temperature (e.g., 25 degrees Celsius) by a temperature- regulation device such as a Peltier device 314. The heat from the Peltier device 314 is extracted through a copper block 312 or other thermally conductive component.

The external surface of the copper block 312 is attached to a heat extraction or heat transfer block 310, connected via one or more heat pipes 308 (six in some embodiments) to a highly efficient heat sink assembly 302. The heatsink assembly 302 (in cooling mode) has two fans 304,306, one on each side to extract the heat. The velocity of the fans 304,306 is controlled by the heating/cooling electronics, which in some embodiments are included in the X-ray detector and electronics block 158. In cold environments the Peltier drive 314 is reversed to heat the detector tiles 155. This arrangement may in some embodiments allow the

apparatus 150 to operate in environments ranging from -35 degrees C to +50 degrees C.

[0083] A fourth potential problem with the tiles 155 is that they are sensitive to shock damage and must be protected externally against this. The environment housing 152 or 318 may be configured in some embodiments to solve this problem as well. In the example shock mounting arrangement 400 shown in FIG. 4, the housing 152 or 318 is attached to the X-ray detector and electronics block 158 using a series of shock absorbing springs to prevent damage should the scanner be dropped or be subjected to high levels of shock.

[0084] FIG. 4A shows a top view of an example shock mounting arrangement 400. The outer environmental housing 318 is attached at housing attachment points 406 to one or more shock absorbing spring suspension units 410, which in turn are attached to a supporting frame for the apparatus 150 at frame attachment points 408. The temperature control system 300 is shown in this view protruding above the housing 318. The shock absorbing spring suspension units 410 serve as a suspension between the supporting frame and the housing 318 to absorb shocks.

[0085] FIG. 4B shows a front view of the same mounting arrangement 400 as FIG. 4A. The housing attachment points 406 can be seen in this embodiment to be angled metal braces or brackets attached along their bottom edge to the

environmental housing 318 and along their angled edge to the shock absorbing spring suspension units 410. The other sides of the shock absorbing spring suspension units 410 attach to the frame attachment points 408, shown here as metal plates. The frame attachment points 408 are configured to attach to a frame suspending the housing 318 such as that shown in FIG. 9-12 and described in detail below (e.g. tubular frame 10 in FIG. 9). [0086] A fifth potential problem with deployment of the tiles 155 in a conventional scanner is that, for many applications, a resolution of 100 microns per pixel is not sufficient. This may be true, for example, on thinner wall pipes being inspected to 'Class B' radiographic standards (such as ISO standard BS EN ISO 17636-2, "Non-destructive testing of welds— Radiographic testing", published by The British Standards Institution, January 31, 2013, which is hereby incorporated by reference in its entirety, hereinafter "IS017636-2" These standards are generally higher than other conventional standards and therefore require higher resolution. For example, ISO IS017636-2 sets out minimum image quality values for metallic materials at Tables B.l to B.14 of that document. As an example potentially applicable to the applications described herein, the minimum values for imaging using the double wall technique, single or double image, with the image quality indicator (IQI) on the detector side are shown in Tables B. ll and B.12 below:

Table B.11— Wire lGl Table B,1 Step and bole IQI

[0087] Duplex IQIs and their application are defined in another standards document, BS EN ISO 19232-5, which is hereby incorporated by reference in its entirety (hereinafter "ISO 19232-5"). A Duplex IQI is defined in ISO 19232-5 as follows:

For measuring image unsharpness, it is a particularly useful toot for establishing and monitoring the performance of radioscopic (realtime) systems. The proposed CEN standard, prEN 13068 will make its use mandatory in certain cases. The IQI consists of 13 wire pairs embedded in rigid plastic. The wires of platinum and tungsten and are exactly spaced to correspond to the diameter of each pair. The degree of unsharpness is indicated by the number of wire pairs that can be seen. As unsharpness increases, the wires merge to form a single image. Each IQI is engraved with a unique serial number and is supplied with a Declaration of

Conformity together with instructions in a storage box.

The ISO 19232-5 model is an exact replacement for the old BS 3971 model MIA, itself identical to CERL B and EN 462-5. This is the only model of duplex IQI in the ISO 19232-5 series.

[...]

13D 0.10 0.030

12D 0.13 0,063

1 1 D 0.16 0.080

10D 0.20 0,100

9D 0.26 0.130

8D 0.32 0.160

7D 0.40 0,200

60 0,30 0,230

5D 0.64 0.320

4D 0.80 0.400

3D 1 ,00 0,500

2D 1 .26 0.630

1 D 1 .60 0,800

[0088] ISO 19232-5 defines the types and positions of image quality indicators as follows:

The quality of image shall be verified by use of image quality indicators (IQIs) in accordance with ISO 19232-5 and ISO 19232-1 or ISO 19232-2. Following the procedure outlined in Annex C, a reference image is required for the verification of the basic spatial resolution of the digital detector system. The basic spatial resolution or duplex wire value shall be determined to verify whether the system hardware meets the requirements specified as a function of the penetrated material thickness in Tables B.13 and B.14. In this case, the duplex wire IQI shall be positioned directly on the digital detector. The use of a duplex wire IQI (ISO 19232-5) for production radiographs is not compulsory. The requirement for using a duplex wire IQI additionally to a single wire IQI for production radiographs may be part of the agreement between the contracting parties. For use on production radiographs, the duplex wire IQI shall be positioned on the object. The measured basic spatial resolution of the digital image (SR_b'^ma9e) (see Annex C), shall not exceed the maximum values specified as a function of the penetrated material thickness (Tables B.13 or B.14). For single image inspection, the single wall thickness is taken as the penetrated material thickness. For double wall double image inspection (Figures 11 or 12), with the duplex wire on the source side of the pipe, the penetrated material thickness is taken as the pipe diameter for

determination of the required basic spatial resolution (SR_b ^ima9e) from Tables B.13 and B.14. The basic spatial resolution of the detector (SR_b ^detector) for double wall double image inspection shall correspond to the values of Tables B.13 and B.14 chosen on the basis of twice the nominal single wall thickness as the penetrated material thickness.

If the geometric magnification technique (see 7.7) is applied with v > 1,2, then the duplex wire IQI (ISO 192325) shall be used on all production radiographs.

The duplex wire IQI shall be positioned tilted by a few degrees (2° to 5°) to the digital rows or columns of the digital image. If the IQI is positioned at 45° to the digital lines or rows the obtained IQI number shall be reduced by one.

The contrast sensitivity of digital images shall be verified by use of IQIs, in accordance with the specific application as given in Tables B.l to B.12 (see also ISO 19232-1 or ISO 19232-2). The single wire or step hole IQIs used shall be placed preferably on the source side of the test object at the centre of the area of interest on the parent metal beside the weld. The IQI shall be in close contact with the surface of the object. Its location shall be in a section of uniform thickness characterized by a uniform grey value (mean) in the digital image.

According to the IQI type used, cases a) and b) shall be considered.

a) When using a single wire IQI, the wires shall be directed perpendicular to the weld and its location shall ensure that at least 10 mm of the wire length shows in a section of uniform grey value or SNR_N, which is normally in the parent metal adjacent to the weld. For exposures in accordance with 7.1.6 and 7.1.7, the IQI can be placed with the wires across the pipe axis and they should not be projected into the image of the weld.

b) When using a step hole IQI, it shall be placed in such a way that the hole number required is placed close to the weld.

For exposures in accordance with 7.1.6 and 7.1.7, the IQI type used can be placed either on the source or on the detector side. If the IQIs cannot be placed in accordance with the above conditions, the IQIs are placed on the detector side and the image quality shall be determined at least once from comparison exposure with one IQI placed at the source side and one at the detector side under the same conditions. If filters are used in front of the detector, the IQI shall be placed in front of the fitter.

For double wall exposures, when the IQI is placed on the detector side, the above test is not necessary. In this case, refer to the correspondence tables ( Tables B.9 to B.14).

Where the IQIs are placed on the detector side, the letter F shall be placed near the IQI and it shall be stated in the test report.

The identification numbers and, when used, the lead tetter F, shall not be in the area of interest, except when geometric configuration makes it impractical. If steps have been taken to guarantee that digital radiographs of similar test objects and regions are produced with identical exposure and processing

techniques, and no differences in the image quality value are likely, the image quality need not be verified for every digital radiograph. The extent of image quality verification should be subject to agreement between the contracting parties.

For exposures of pipes with diameter 200 mm and above with the source centrally located at least three IQIs should be placed equally spaced at the circumference.

The IQI images are then considered representative for the whole circumference.

[0089] Once again, the skewed configuration of the X-ray sensitive area 154 may address this potential problem. In applications where a higher resolution and a better signal-to-noise ratio (SNR) are required, the detector skew can be used to increase the resolution of the image. In some embodiments, this technique involves extracting data from adjacent pixels in the pixel array (e.g. the 768 x 256 pixel array) and processing them in such a way as to increase the resolution both across the weld and in the scan direction 162. Rotating or skewing the X-ray sensitive area 154 of the detector apparatus 150 introduces a sub-pixel shift of each row of pixels to the next, perpendicular to the direction of motion. Combining the data from successive samples of the same region allows spatial oversampling in this direction, with a corresponding increase in resolution. In some embodiments, the technique applied for performing this combination is to increase the number of pixels in the image using interpolation, as part of the digital image rotation before doing TDI.

[0090] In some embodiments, the image of the weld may be blurred in the direction of motion (i.e. the scan direction 162) due to the distance the apparatus 150 travels during each frame acquisition. This may further limit the spatial resolution of the TDI image in this direction. Increasing the frame rate may reduce this blur. Increasing frame rate to greater than one frame for each one-pixel (e.g. 100 micron) motion over the weld causes a sub-pixel shift of one frame to the next. Combining the data from successive frames allows spatial oversampling in this direction, with a corresponding increase in resolution. This combination of data can be performed in a similar manner to the TDI techniques described above for combination and interpolation of adjacent pixel values.

[0091] FIG. 7 is a flowchart showing an example image enhancement method 700 for increasing the resolution of scanned image data for an object (such as a pipe weld) by combining adjacent pixel data collected by a skewed array of sensor tiles (such as X-ray sensitive tiles 155 used as the X-ray sensitive area 154 of scanning apparatus 150, with a longitudinal axis of the array along which the tiles 155 are arranged being skewed relative to a scan direction 162). Steps 602 through 610 of this method are identical to those of the TDI method 600 described above, with only the final step 712 of combining the image datasets being different.

[0092] In this method 700, the combination function applied to the image datasets combines pixels corresponding to adjacent, overlapping locations (rather than pixels corresponding to the same scanned location in step 612 of TDI method 600). The combination function used in step 712 would generally be different from the combination functions used in method 600 - to generate a single pixel of the output image, the present described method 700 would in some embodiments combine one or more pixels from each dataset to generate a weighted average of the pixel value corresponding to a location on the scanned object, with each dataset pixel being weighted based on its degree of overlap with the target location. In an example first dataset, sub-pixel displacement relative to a second dataset means that a single pixel in the first dataset would correspond to multiple pixels in the second dataset that all overlap the location covered by the single pixel from the first dataset. Thus, the combination function may combine the single pixel from the first dataset with the multiple pixels from the second dataset, with the single pixel from the first dataset being weighted more heavily than each of the multiple pixels from the second dataset because they only overlap the target location partially instead of fully.

[0093] A skilled person will appreciate that a number of sub-pixel combination algorithms, including interpolation algorithms, may be used to enhance image resolution using multiple image datasets containing sub-pixel displacements between datasets.

[0094] In some embodiments, the sub-pixel displacements giving rise to this overlap between adjacent pixels in time-adjacent scans is solely a function of the skew of the array. In other embodiments, oversampling (i.e. moving the scanner a distance between scans that is less than the distance corresponding to one pixel) may also give rise to further sub-pixel displacement between scans. For example, in some embodiments wherein 100 microns corresponds to one pixel, 1.5 to 2 frames may be collected per 100 microns of scan distance.

[0095] These various techniques may be combined in some embodiments. When applied either separately or together, they may improve the SNR and/or the image resolution across the weld and/or in the weld scan direction 162 relative to conventional scanning, with a relatively small decrease in the weld inspection speed. Even a moderate skew angle of 5 degrees and oversampling by 2: 1 may increase the resolution by up to 30% in some applications. This results in the 100- micron pitch basic array (768 x 256 pixels) yielding 70 microns per pixel effective resolution.

[0096] Another benefit of these techniques is that that they have the potential to reduce pixilation and aliasing effects. FIG.s 14A and 14B (illustrated by line drawings FIG. 8A and 8B respectively) show example images from an X-ray scanning apparatus without these pixel enhancement techniques (FIG. 14A) and from an example embodiment of the scanning apparatus 150 using these pixel enhancement techniques (FIG. 14B). The images are duplex image quality indicator (IQI) images obtained from a scan of a 15mm thick wall using a double-wall, single-image (DWSI) scan. In FIG. 14A, the image displays significant visible pixilation and aliasing as well as a relatively large amount of noise (and hence a low SNR). In particular, the lead (Pb) number "35" displays substantial pixilation. In contrast, in FIG. 14B, the image displays no notable visible pixilation or aliasing and has very little visual noise (i.e. a higher SN R). [0097] Standards document ISO 176363, discussed above, describes techniques for calculating SNR. SNR and normalized SNR (SN R_N) are defined therein as follows:

3.10

signal-to-noise ratio

SNR

ratio of mean value of the linearized grey values to the standard deviation of the linearized grey values (noise) in a given region of interest in a digital image

3.11

normalized signal-to-noise ratio SNRN

signal-to-noise ratio , SNR, normalized by the basic spatial resolution, SR_b, as measured directly in the digital image and/or calculated from the measured

SNR, SNRmeasured_/ by

SNRN ⁼ SNRmeasured 88,6 / SRb

[0098] Using these calculations, the test image shown in FIG. 14B has a SNR of 155 compared to SNR 72 for the image in FIG. 14A. This image enhancement was achieved by sampling at 2 frames per 100 micron of movement. Ordinarily, one would expect this over-sampling to improve the SNR of the FIG. 14A image by the square root (N frames per pixel-length of 100 microns, N = 2, SNR should improve to the power of 2) from 72 to 102. However, in FIG. 14B duplex pair D10 812 is clearly split, whereas the first merged duplex pair is Dll 814 (i.e. that is the first duplex pair of lines that do not appear as two distinct lines). In contrast, the first merged duplex pair in FIG. 14A is D10 802, with aliasing being visible on D9 804 and below. Accordingly, the SNR result for FIG. 14B is 155 using the calculation methods described in ISO 176363. This should mean that these image

enhancement techniques can be achieved with no loss of scan speed in some embodiments.

[0099] By combining these various described features and techniques, example embodiments may overcome these various potential problems or complications and provide an X-ray detection system or apparatus 150 with no boundary effects or missing data. The apparatus 150 may have an ambient temperature operating range of from -35 degrees C to +50 degrees C and may be able to withstand high level shocks. The mechanical configuration and image processing algorithms may eliminate detector artefacts and also increase the resolution from the native 100 microns/pixel down to 50 microns/pixel.

[00100] Furthermore, the slow scanning time of conventional systems described in the Background section above - up to one hour to scan a 36" heavy wall pipeline girth weld - may in some embodiments be reduced by eight-fold for similar image quality. At slower speeds, both resolution and contrast may be improved considerably in the apparatus 150 relative to conventional systems.

[00101] Thus, the present disclosure in some embodiments provides high- performance X-ray detection apparatus, methods and systems for applications including pipeline weld inspection. At least four major areas of innovation are described herein : first, a highly sensitive X-ray photon counting technology in the form of the X-ray sensitive modules or tiles 155 described above. Second, a methodology to eliminate artefacts produced by detection modules boundaries and data loss from detector module gaps by applying a skew to the modules/tiles 155 and suitable image processing. Third, an environmental protection system to enable the X-ray detector tiles and electronics to be kept at an optimal temperature, such as 25 degrees Celsius, while the external ambient temperature can typically vary, e.g. from -35 degrees C to +50 degrees C. The housing may also include an anti- shock mounting to protect the fragile tiles 155. Fourth, a methodology to increase both the basic resolution and signal to noise by data extraction from the detector module 'skew' and over-sampling. Each of the second through fourth features addresses one or more potential problems that may arise from deploying the first feature in a scanning application.

[00102] Deployment of example embodiments described above may be in various X-ray detection applications, including pipe weld scanning. Example supporting structures for an example X-ray scanning apparatus 150 will now be described in the context of pipe weld inspection. [00103] FIG. 9 shows an example X-ray detection apparatus 3 mounted to a pipe by means of a track 7. The X-ray detection apparatus 3 includes an X-ray detector 1 and an electronics module 2 for collecting the image data from the X-ray detector 1 and communicating with a data processing module (not shown). The X- ray detection apparatus 3 is contained within a tubular frame 10, which may include a liquid cooling system. In other embodiments, the X-ray detection apparatus 3 is contained within an environmental enclosure as described in greater detail above.

[00104] The track 7 is configured to encircle a pipe, such as a metal pipe used in a pipeline. Securing devices 14 are used to clamp or otherwise secure the track 7 to the pipe, with mounting pads 18 used to space the track7 away from the pipe to allow traversal by the X-ray detection apparatus 3. The X-ray detection apparatus 3 is driven around the pipe on the track 7 by means of one or more actuators, in this embodiment shown as a first motorized buggy 4.

[00105] Situated on the track 7 on the opposite side of the pipe from the X-ray detection apparatus 3 is an X-ray source 5 driven around the track by a second motorized buggy 6. In operation, the two buggies 4,6 move along the track 7 to maintain the opposing positions of the X-ray detection apparatus 3 and the X-ray source 5 as the X-ray detector 1 scans the pipe.

[00106] FIG. 15 shows a photograph of a further example embodiment of a skewed X-ray scanning apparatus 1400 deployed on a pipe 1450 for scanning of a weld 1452 on the pipe 1450. This apparatus 1400 traverses the circumference of the pipe 1450 on a band 1440. The apparatus 1400 uses an actuator, implemented here as a motor 1404, to move itself around a gear-toothed track 1442 of the band 1440. The X-ray sensitive array 157 of the apparatus 1400 is housed inside an environmental housing 318 as described above. Electronics for data collection and motor control, as well as a power supply, are housed within an electronic housing 1406. Conduits 1402 lead away from the electronic housing 1406 to, e.g., the data collection module 170 (not shown) in the form of a conventional desktop or laptop computer. The heat sink and fan of a temperature control system 300 are visible protruding from the housing 318 on the left side of the apparatus 1400. A shock absorbing spring suspension unit 410 is also visible suspending the environmental housing 318 below the rest of the apparatus 1400.

[00107] FIG. 16 is a further photograph of a front view of a similar apparatus 1400 to that shown in FIG. 15. An X-ray source 5 is situated on the band 1440 on the opposite side of the pipe 1450 from the apparatus 1400.

[00108] FIG. 17 is a further photograph of a close up front view of a similar apparatus 1400 to that shown in FIG. 15 and FIG. 16.

[00109] The core X-ray detection in this example embodiment is implemented using a heavily customized photon counting detection modules and their associated read-out electronics. This technology uses both photon counting and a large sensitive area to achieve a sensitivity almost 8 times higher than other known techniques.

[00110] The detector units are made up of 6 modules (i.e. tiles 155) of 128 x 256 pixels each, with a module gap 203 of 100 microns, covering an active width of 77mm. Alternatively, 8 modules can be accommodated within the inner housing. To avoid any data loss and visible boundary gaps at the module boundaries, the detection modules are mounted in the environmental housing 318, typically with a 5 degree skew. This skew allows the software running on the data processing module 170 to eliminate module boundaries from degrading the image, as described in detail above.

[00111] The software then allows the overall 768 (or 1024) pixel wide x 256 pixel high sensitive area to be read out as single frames, corrects each frame for bad pixels/non-uniformity, de-skews each frame, and then sums 256 rows for every 100 microns of movement as the detector is scanned over the weld surface using TDI. This produces a single line of data for display to an operator. This embodiment is able to achieve 500 frames/second read-out, which equates to a weld scan speed of 50mm/second. [00112] The environmental housing 318 protects the internal detector from the external environment, and the temperature control system 300 (implemented as a Peltier air cooling/heating system as described above) has been tested from -35 degrees C to + 50 degrees C.

[00113] The apparatus 1400, including the cooling system 300 and other components mounted on the wide band 1440, weighs less than 30 kg in this embodiment.

[00114] The example scanning apparatus 1400 has a native resolution of 100 microns/pixel, which is sufficient for weld inspection of wall thicknesses to 'Class B' Standards down to Vi" wall. Generally, a 100-micron pitch detector has an un- sharpness of > 0.2mm, so can only reliably split duplex pair D9 of the IQI defined in EN 462-5 (0.26mm Ug) at 20% modulation depth. This makes duplex D10 the first merged pair as defined in ISO 176363.

[00115] The configuration of the scanning apparatus 1400 implements the techniques described above to improve scanned image resolution, improve signal to noise ratio, and reduce pixilation and aliasing. In its standard data collection mode, the apparatus 1400 outputs one frame for every 100 microns of movement over the weld. For example, scanning a weld at lOmm/second will generate 100 frames/second of data.

[00116] Tests using the example apparatus 1400 have shown that resolution can be increased by frame oversampling and pixel interpolation. Oversampling and interpolation of the skewed detector output have been used to improve the un- sharpness from 0.26mm to 0.2mm or better. This improved resolution allows Duplex D10 to be split, making Dll the 1st merged pair. Reduced pixilation and aliasing both make the overall image easier to interpret, and oversampling and interpolation appear to improve the SNR.

[00117] These results were tested by generating image data under a number of different conditions. A set of short weld images were collected at higher frame rates than the default 100 microns/frame, covering the range 50 microns/frame to 100 microns/frame. The data processing module 170 in this example uses algorithms developed in the Python programming language and using the open source image processing software OpenCV (Open Source Computer Vision Library) to take advantage of the oversampled frame data and the unique configuration of the apparatus 1400.

[00118] Based on these tests, it is predicted that the example apparatus 1400 may be able to achieve Class B results on single wall weld thicknesses down to 10mm or less. Image data may be produced having overall resolution similar to that of existing techniques, but at eight times the speed.

[00119] The embodiments of the present disclosure described above are intended to be examples only. The present disclosure may be embodied in other specific forms. Alterations, modifications and variations to the disclosure may be made without departing from the intended scope of the present disclosure. While the system, devices and processes disclosed and shown herein may comprise a specific number of elements/components, the systems, devices and assemblies could be modified to include addition or fewer of such elements/components. For example, while any of the elements/components disclosed may be referenced as being singular, the embodiments disclosed herein could be modified to include a plurality of such elements/components. Selected features from one or more of the above-described embodiments may be combined to create alternative embodiments not explicitly described. All values and sub-ranges within disclosed ranges are also disclosed. The subject matter described herein intends to cover and embrace all suitable changes in technology. All references mentioned are hereby incorporated by reference in their entirety.

Claims

1. An X-ray detection apparatus for scanning a scanned object along a scan direction using X-ray detection, comprising :

an X-ray detector and electronics module comprising :

an X-ray sensitive area for detecting X-rays passing through or reflected from the scanned object, comprising a plurality of X-ray sensitive tiles, each X-ray sensitive tile comprising a two-dimensional array of pixels, the plurality of X-ray sensitive tiles being arranged in a linear array having a width dimension and a height dimension perpendicular to the width dimension, the height dimension being skewed relative to the scan direction by a skew angle, the skew angle being greater than zero degrees and less than forty-five degrees; and

electronics for collecting X-ray image data from the X-ray sensitive area.

2. The apparatus of claim 1, further comprising an actuator for driving the apparatus in the scan direction.

3. The apparatus of claim 2, wherein each of the plurality of X-ray sensitive tiles comprises a crystal array comprising either Cadmium Telluride (CdTe) or Cadmium Zinc Telluride (CdZnTe).

4. The apparatus of claim 3, wherein each of the plurality of X-ray sensitive tiles further comprises a complementary metal-oxide-semiconductor 2-dimensional array bump-bonded to the crystal array.

5. The apparatus of claim 4, further comprising an environmental housing for protecting the X-ray detector and electronics module from environmental hazards.

6. The apparatus of claim 5, further comprising a temperature control system for controlling the temperature inside the environmental housing, and wherein the environmental housing is insulated.

7. The apparatus of claim 6, wherein the temperature control system comprises a Peltier device.

8. The apparatus of claim 5, further comprising a shock absorbing suspension coupled between the X-ray detector and electronics module and the environmental housing for protecting the X-ray detector and electronics module against physical shocks.

9. The apparatus of claim 4, wherein each X-ray sensitive tile comprises a rectangular matrix of pixels having a width dimension aligned with the width dimension of the linear array and a height dimension aligned with the height dimension of the linear array.

10. The apparatus of Claim 9, wherein the plurality of X-ray sensitive tiles are arranged in the linear array with a gap between each pair of adjacent tiles.

11. The apparatus of Claim 10, further comprising a data processing module for receiving X-ray image data from the electronics.

12. The apparatus of Claim 11, further comprising a controller for:

perform a single scan by controlling the X-ray detector and electronics module to collect X-ray image data and provide this data to the data processing module; and controlling the actuator to move the apparatus a predetermined between-scan distance in the scan direction after each scan.

13. The apparatus of Claim 12, wherein the data processing module is configured to:

receive a first scan dataset from the X-ray detector and electronics module corresponding to the X-ray image data collected from a first scan;

receive one or more additional scan datasets from the X-ray detector and

electronics module corresponding to the X-ray image data collected from one or more scans following the first scan;

de-skew the first scan dataset and the one or more additional scan datasets; and apply time delay integration to the first scan dataset and the one or more additional scan datasets to generate an output image.

14. The apparatus of Claim 13, wherein :

a portion of the output image corresponding to a portion of the scanned object is generated by applying time delay integration to a predefined number of datasets containing X-ray image data corresponding to the portion of the scanned object; and

applying time delay integration to generate an output image comprises:

once the predefined number of datasets corresponding to a first portion of the scanned object have been received by the data processing module, generating a first portion of the output image corresponding to the first portion of the scanned object by applying time delay integration to the predefined number of datasets; sending the first portion of the output image to a user device; and

repeating the steps of generating a portion of the output image and sending the portion of the output image to the user device for each subsequent portion of the output image as subsequent datasets are received by the data processing module.

15. The apparatus of Claim 14, wherein the portion of the output image is a one- pixel-wide line of the output image.

16. The apparatus of Claim 12, wherein the between-scan distance is less than a distance corresponding to a single pixel height in the X-ray image data,

and wherein the data processing module is configured to:

receive one or more additional scan datasets from the X-ray detector and

de-skew the first scan dataset and the one or more additional scan datasets; and apply time delay integration to the first scan dataset and the one or more additional scan datasets to generate an output image,

wherein de-skewing a dataset comprises interpolating between adjacent pixels in the dataset to increase the number of pixels in the dataset.

17. The apparatus of claim 4, wherein the electronics are configured to segment the X-ray image data into a plurality of energy bands.

18. The apparatus of claim 4, wherein the electronics are configured to filter image noise in the X-ray image data based on an energy threshold.

19. A method for scanning a scanned object along a scan direction using X-ray detection, comprising :

applying an X-ray source to the scanned object;

detecting X-rays passing through or reflected from the scanned object using a plurality of X-ray sensitive tiles, each X-ray sensitive tile comprising a two- dimensional array of pixels, the plurality of X-ray sensitive tiles being arranged in a linear array having a width dimension and a height dimension perpendicular to the width dimension, the height dimension being skewed relative to the scan direction by a skew angle, the skew angle being greater than zero degrees and less than forty-five degrees;

collecting X-ray image data from the X-ray sensitive tiles;

moving the X-ray sensitive tiles a predetermined distance along the scan direction; and

repeating the steps of applying, detecting, collecting, and moving until a region of interest of the scanned object has been scanned.

20. The method of claim 19, wherein each of the plurality of X-ray sensitive tiles comprises a complementary metal-oxide-semiconductor 2-dimensional array bump- bonded to a crystal array comprising either Cadmium Telluride (CdTe) or Cadmium Zinc Telluride (CdZnTe).